LU

Lumosia MoE 4x10.7

LumosiaGGUF by TheBlokeUpdated 2024-01-10
Download on Hugging Face
Parameters36B
Context4K
Layers48
Architecturellama
Smallest quant12.0 GB
Quick answer

Lumosia MoE 4x10.7 needs about 22.5 GB of memory at Q4_K_M (a 20.4 GB file) with an 8K context window, so it fits on a 24 GB GPU. This repository has 12 quants, from 12.0 GB (Q2_K) to 38.4 GB (Q8_0).

Lumosia MoE 4x10.7 – a 36B parameter mixture of experts for chat and roleplay

Lumosia MoE 4x10.7 is a 36B parameter model built as a mixture of experts from several Solar models. It supports a 4096 token context and is licensed under apache-2.0. The model is available in multiple GGUF quantisations.

Good at

  • Chat and roleplay
  • Logical reasoning
  • Coherent long‑form text

Limits

  • No vision support
  • Requires 8k context for memory estimates
  • Experimental model

Best for

  • Chatbot development
  • Story generation
  • Roleplay scenarios

Who it’s for. Developers and hobbyists looking for a large, open‑licensed LLM that can handle roleplay and logical tasks. The model is experimental and may need tuning for production use.

Quants & memory

QuantFileEst. memoryvs VRAMResultDownload
Q2_K12.0 GB 13.3 GB
Too big Get file →
Q3_K_S15.6 GB 16.9 GB
Too big Get file →
Q3_K_M15.7 GB 17.0 GB
Too big Get file →
Q3_K_L15.8 GB 17.1 GB
Too big Get file →
Q4_020.3 GB 21.6 GB
Too big Get file →
Q4_K_S20.3 GB 21.7 GB
Too big Get file →
Q4_K_M20.4 GB 21.7 GB
Too big Get file →
Q5_024.8 GB 26.1 GB
Too big Get file →
Q5_K_S24.8 GB 26.1 GB
Too big Get file →
Q5_K_M24.9 GB 26.2 GB
Too big Get file →
Q6_K29.6 GB 30.9 GB
Too big Get file →
Q8_038.4 GB 39.7 GB
Too big Get file →

For 8 GB VRAM, Q4_K or Q4_K_S (20.34 GB file, ~22.5 GB memory at 8k context) is suitable. For 16 GB VRAM, Q5_K_M (24.85 GB file, ~27 GB memory) offers better quality. Higher quants like Q6_K or Q8_0 require more than 32 GB VRAM.

What GPU do you need for Lumosia MoE 4x10.7?

These figures add the file size, an fp16 KV cache at an 8K context window (worked out from this model’s own header) and about 0.5 GB for the runtime. They are estimates, so leave a little headroom, and expect to use less context or a smaller quant if you are right at the limit.

  • 4 GB VRAM: no quant fits at an 8K context.
  • 6 GB VRAM: no quant fits at an 8K context.
  • 8 GB VRAM: no quant fits at an 8K context.
  • 12 GB VRAM: no quant fits at an 8K context.
  • 16 GB VRAM: the best quant that fits is Q2_K (12.0 GB file, about 14.1 GB in use).
  • 24 GB VRAM: the best quant that fits is Q4_K_M (20.4 GB file, about 22.5 GB in use).
  • 32 GB VRAM: the best quant that fits is Q6_K (29.6 GB file, about 31.7 GB in use).
  • 48 GB VRAM: the best quant that fits is Q8_0 (38.4 GB file, about 40.5 GB in use).

Which Lumosia MoE 4x10.7 quant should you choose?

Q4_K_M is the usual starting point for Lumosia MoE 4x10.7: it balances file size and quality well. Here it is a 20.4 GB file.

The Q8_0 file is 38.4 GB, about 1.9 times the size of Q4_K_M. It keeps the weights closer to the original, which only matters if you have the memory to spare.

If memory is tight, the smallest file is Q2_K at 12.0 GB, with the quality trade-offs that come with fewer bits. Use it only when memory is very tight. Expect a clear drop in quality, which is usually worse on small models and easier to live with on large ones.

New to quant names? Read GGUF quantization explained.

Specifications

Architecture from the GGUF header

Architecturellama
Parameters36B (approximate, mixture of experts)
Layers48
Trained context4,096 tokens
KV cache (fp16)192 KB per token

Source

Quantized byTheBloke
FamilyLumosia
Licenseapache-2.0
Downloads (HF)1,215
Last updated2024-01-10

Frequently asked questions about Lumosia MoE 4x10.7

How much VRAM does Lumosia MoE 4x10.7 need?

About 22.5 GB at Q4_K_M with an 8K context window. That is the 20.4 GB file, plus the KV cache and roughly 0.5 GB of runtime overhead. Smaller quants need less and longer context needs more.

Can I run Lumosia MoE 4x10.7 on an 8 GB GPU?

Not at an 8K context. The smallest file needs about 14.1 GB, so you would need a larger GPU or CPU offloading.

What is the file size of Lumosia MoE 4x10.7 GGUF?

It depends on the quant: Q2_K is 12.0 GB, Q3_K_S is 15.6 GB, Q3_K_M is 15.7 GB, Q3_K_L is 15.8 GB, Q4_0 is 20.3 GB, Q4_K_S is 20.3 GB and 6 more.

What context length does Lumosia MoE 4x10.7 support?

The GGUF header lists a trained context length of 4,096 tokens (4K). The KV cache grows with the context you use: about 1.6 GB at 8K and 0.8 GB at 4K.

Where can I download Lumosia MoE 4x10.7 GGUF?

The files are on Hugging Face in the TheBloke/Lumosia-MoE-4x10.7-GGUF repository. The download buttons on this page open the file on Hugging Face. ggufmodels.com does not host any model files.

What license does Lumosia MoE 4x10.7 use?

Hugging Face lists the license as apache-2.0. Licenses can differ between the quantized files and the original model, so check the repository and the base model before commercial use.

Can I run Lumosia MoE 4x10.7 on 8 GB VRAM?

Use the Q4_K or Q4_K_S quant, which estimates 22.5 GB memory at 8k context – this exceeds 8 GB, so it is not feasible without further optimisation.

Does the model support vision tasks?

No, the model does not have vision capabilities.

What license does the model use?

The model is released under the apache-2.0 license.

Other GGUFs of Lumosia-MoE-4x10.7

Similar models

Browse all →

More from TheBloke

All by TheBloke →

More Lumosia models

All Lumosia →

Other large models

All large models →